You are on page 1of 13

The UTS Durability Test for Earth Wall Construction.

By Dr. Kevan Heathcote & Gregory Moor University of Technology Sydney

This paper looks at the development of earth buildings in Australia and examines the Bulletin 5 accelerated erosion test which was introduced in the 70¶s to deal with the question of durability. The paper then goes on to outline the limitations of this approach and gives details of a new durability spray test developed by the authors at the University of Technology, Sydney.

Durability, earth walls, erosion, spray test

To examine the issues effecting the durability of earth walls in Australia we must appreciate the historical and environmental context of earth building on this continent.

Historical Context Australia has a short European based building history. The country was first settled by the English in 1788 and the early buildings were crude constructions and of temporary nature. There are no remaining earth buildings before the mid nineteenth century, however we have written reference to the use of earth wall construction during this early period of settlement. The first European settlers who arrived in Sydney Cove were not aware of Aboriginal construction methods but soon found the small acacia trees were suitable for wattling and plastering with clay.


The earliest remaining earth buildings are from the mid nineteenth century and are all located in rural areas far from the city. built of wattles. (note the wide roof overhang and rendered walls) Environmental Context With the advent of the industrial revolution earth wall construction declined during the mid nineteenth to mid twentieth century. Termites.The trees became known as wattles and the building process wattle and daub. It should be noted that these remaining buildings have well maintained surface coatings as well as wide roof overhangs (Figure 1) Figure 1 Pise¶ house near Bathurst c1850. rain and increasing property development led to the destruction of all early earth wall buildings in Sydney. The preferred wall 2 . plastered with clay and thatched. however shortages of building materials following the Second World War led to a renewed interest in earth wall construction. Governor Philip began a new settlement at Parramatta and before the end of 1790 there were thirty -two houses completed.

The traditional areas of earth building such as Egypt and North Africa have annual rainfalls averaging between 200mm and 500mm per material in Australia is fired clay brick without an external coating and this same appearance is preferred in earth wall construction. Although Australia is a very dry continent the majority of the population live in dwellings in the coastal regions with annual rainfall averaging between 600mm and 1800mm per year (Figure 2). In the old villages the 3 . The erosion of earth walls by wind driven rain coupled with the preference for uncoated surfaces has led to studies on the durability of earth walls in the Australian environment. ³We note also that in the United Kingdom and France that earth walling is limited to the smaller domestic and farm buildings. Figure 2 Australian Annual Rainfall Majority of dwellings are in the 600 to 1800 areas. Due to their limited durability in an unstabilised state earth buildings have in the past been seen to be inferior to more permanent materials such as stones and fired clay bricks.

are protected by large overhanging eaves. What these experiments demonstrated. and any buildings having more considerable architectural pretensions. 4 . p5) The perceived lack of durability of earth has been a significant barrier to its acceptance as a modern building material. This test is referred to as the ³Bulletin 5´ accelerated erosion test (Figures 3 & 4) as that is the name of the document in which it is contained. 1958. 1981). Major earth buildings that have survived over long periods are mainly located in areas of minimal annual rainfall.parish church and the manor house. was the dramatic effect climatic conditions have on the durability of earth walls. or are covered with protective coatings. Thus we may take as a tacit admission that unstabilised earth walling did not possess sufficient permanence to justify the expenditure of a large amount of effort and elaboration in fittings and decorative work.1998). we re invariably built of brick or stone. For effective prediction of the service life of earth buildings it is necessary to have an accelerated durability test which is a reliable predictor of in -service performance. Bulletin 5 Spray Test In response to an increased interest in earth construction in the 70's the Commonwealth Experimental Building Station in Australia developed an accelerated erosion test based on spraying water horizontally onto specimens using a specific nozzle (Schneider. after 43 years of exposure. This spray test is called up in the Building Code of Australia and a modified version was included in the New Zealand Code of Practice on earth wall buildings (NZS 4297.´ (Fitzmaurice. Middleton (1952) constructed many rammed earth test walls at the Commonwealth Experimental Building Station in Sydney in 1949.

The total depth after one hour is div ided by sixty to give erosion in mm per minute. The test is interrupted at fifteen -minute intervals and the depth of erosion measured with a 10mm diameter flat -ended rod. 5 .Figure 3 Bulletin 5 Accelerated Erosion Test Figure 4 Bulletin 5 Accelerated Erosion Test The test consists of spraying the face of a sample for a period of one hour or until the sample is penetrated. The maximum permissible erosion rate for all types of earth construction is one mm per minute.

this being similar to recorded values of wind velocity during rain in Sydney. The first is that the test does not in any way simulate rainfall and as can be seen in Figure 5 actually bores holes in the specimens. In the end the Fulljet series of nozzles (manufactured by Spraying Systems Company in Illinois) were found to be ideal in that they produce a narrow spray which is made turbulent by the internal vanes (See F igure 6) and were relatively inexpensive. 6 .Development of UTS Spray Test Two general difficulties in using the Bulletin 5 test to predict earth wall performance became obvious to the authors. Figure 5 Earth Block showing erosion created by Bulletin 5 Test The authors investigated many different nozzles to make the basic Bulletin 5 spray test setup more representative of the turbulent erosion pattern of rainfall. Secondly in interpreting the results of the test no consideration is given to the climatic conditions in which a proposed building is to be located. The 1550 nozzle was chosen for its ability to produce stream velocities of around 9 m/sec.

WIND-DRIVEN RAIN (knot mm) 4000 NW 3000 N NE 2000 1000 W 0 E Series D SW S SE Figure 7 Wind Driven Rain Rose (Heathcote.2003) A testing procedure was developed where the one specimen was tested in the laboratory and then in the field.2003). These tests involved measuring the wind driven rainfall at the site (See Figure 7) and correlating the ero sion of the field specimens relative to the erosion of the laboratory specimens adjusted for the relative volumes of water impacting the specimens (Heathcote. The field -testing was done at a weather station where wind and rainfall data was available.Figure 6 FullJet Full Cone Nozzle Field tests were then conducted by the authors over a three year period to establish a relationship between erosion in the field and erosion in the laboratory using the new setup. From previous experience it was noted that wind driven rain predominantly came from the south and 7 .

Based on established relationships between erosion and water velocity correction factors were then established for situations where the average wind speed during rain was higher or lower than 7 m/sec. assuming an average wind speed during rain of around 7 m/sec. Figure 8 Specimens being Field Tested at Weather Statio n The results of this investigation enabled the authors to produce a relationship between the annual rainfall at a particular site and the spraying time necessary for there to be a one to one relationship between the erosion depth in the field and the ero sion depth of specimens in the laboratory. 8 .therefore the specimens were faced in a north/south orientation to receive maximum exposure. and a service life of 50 years. Details of this work are yet to be published but are based on the work of Heathcote (2003). To simulate wall construction the edges were protected from the weather by a PVC tube (Figure 8).

FULLJET 1550 NOZZLE 350 mm RUBBER GASKET SPECIMEN 70 kPa 100 DIA HOLE igure 9 pray est etup In the test specimens are placed with their external face surface exposed to the spray. A Fulljet 1550 nozzle is positioned 350 mm from the face and water is sprayed at a pressure of 70 kPa.UTS Spray Test Details of the setup of the UTS spray test are given in Fi gure 9. Figure 10 shows a specimen being tested at UTS.0 where average wind during rain =7 m/sec (Default) = 2. which impacts the specimens through a 100 mm d iameter hole. The time of exposure for the specimens is calculated as follows Time of Exposure (mins) = Annual Rainfall (mm) /10 v Wind Factor Where Wind Factor = 0. The runoff water is filtered before being re -cycled.5 where average wind during rain < 4 m/sec = 1.0 where average wind during rain < 4 m/sec 9 .

For a rainfall of 600 mm in a low wind area the time would be 600/10 v 0. Therefore Predicted Average Loss of Wall Thickness Over 50 Year Period = 2 v Measured Erosion depth Predicted Maximum Localised Loss of Wall Thickness = 1.5 = 30 minutes. 10 . Specimens are then sprayed for the calculated duration and the resulting erosion depth measured.5 v Predicted Average Loss For example if a building is to be located in Sydney and the maxim um depth of erosion after 120 minutes is 5 mm then the predicted average wall thickness loss over 50 years would be 10 mm with localised areas of erosion reaching 15 mm. Furthermore. it is assumed that local areas of erosion 50% greater than that calculated will occur. This erosion depth is an indication of the mean erosion depth to be expected for in service conditions but is to be multiplied by a factor of safety of 2 to take into account the limitations of the experimental data.For example if the mean annual rainfall in Sydney is 1200 mm (Sydney) then the time of exposure is 120 minutes.

and a similarity can be drawn between acceptable levels of erosion and acceptable classes of surface finish in concrete work. bearing in mind that earth walls are generally much thicker than normal masonry walls. Erosion therefore is more a problem of aesthetics. For a 250 thick single storey wall even a loss of 50 mm will have little structural significance. Adopting 3 categories we might define 11 .Figure 10 UTS Accelerated Erosion Test Allowable Wall Erosion Erosion of walls rarely pose a structural problem.

Commonwealth Experimental Building Station.A.A. Heathcote. Faculty of Engineering. United Nations. ³Durability of Earthwall Buildings ´.A. 6. K. Syd ney.F. Heathcote.. ³ Relationship Between Spray Erosion Tests and the Performance of Test Specimens in the Field´. The UTS Erosion test given in this paper has been developed as a result of extensive field and laboratory testing and provides a logical basis for acceptance testing of earth building materials used in a particular climatic region.Construction and Building Materials ³. Fitzmaurice. Spain 2000 . The required surface finish would then be specified by the Client and that would form the basis of the acceptance testing of ma terials in accordance with the UTS test outlined above. No 6. Sustainable Construction into the Next Millenium ± Environmentally Friendly and Innovative Based Materials. ³Manual on Stabilised Soi l Construction for Housing´..1952.A.G.Bulletin No 5.1995 5. Nov 2000. Technical Assistance Program..A. ³Durability of Cement Stabilised Earth Walls´. Heathcote. 1958.J. Heathcote. ³Resistance of Earthwall Buildings to Wea thering by Wind-Driven Rain´. PhD Thesis. Heathcote. Vol 9 No 3.K. 2. Barcelona. ³Durability of Earth Wall Units´. K. References 1.R.. Joao Pessoa. Australia. Class 3 Surface ± Surface where the average surface erosion over a 50 year period is not expected to exceed 12 mm with local areas of erosion of 18 mm.K. Class 2 Surface ± Surface where the average surface erosion over a 50 year period is not expected to exceed 8 mm with local areas of erosion of 12 mm. Middleton. Conclusions Little work has been done to date in developing a laboratory test which is a reliable predictor of the in-service erosion of earth wall buildings. and Sri Ravindrarajah. 3. The Australian Institute of Building Papers . pp13-20.´Earth -Wall Construction´. K.. and Moor.2002. Brazil. University of Technology Sydney. 4. Fifth CANMET/ACI International Conference on Durability of Concrete.Class 1 Surface ± Surface where the average surface erosion over a 50 y ear period is not expected to exceed 4 mm with local areas of erosion of 6 mm.G..1995. 7. pp185-189. R. 12 .

html (30/09/2010) 13 .Durability Research´ Moor. Germany ISBN 3 -81676118-6 From: http://www. Moderner Lehmbau 2002. ³Earth Building in Australia . and Heathcote. Berlin.G.